You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We have had several debates, and discussions about tags, and annotations (any many other forms of the same thing). For the foreseeable year, I believe this argument holds true:
Not that many people used tags or annotations in ADS 2.0, and so they are low priority
We have made it possible for people to export their libraries to disk if they wish to retain tags, and import them into other tools such as Zotero, or Mendeley.
With the above, we have removed any rush or prerequisite to fulfil any requirement such as -- "ADS 2.0 had it, so we must implement it before we shut down".
So, I hesitantly write the following, and encourage that this feature only be considered if the developers and other team members think it is important.
Tags or Annotations: Backend
Given the way the system is currently architected, it is completely possible to include tags or annotations.
Bibcode level
Bibcode level refers to, annotating a single bibcode Bib1 with a tag, or an annotation, within a given library Lib1.
Keep in mind, this would always have a Library scope. If you wanted to use the same tags and annotations for a user through all of our interface, I do not believe the library service is the way to do it, given the way the models are -- but worth thinking about.
Library level
Library level refers to the scope of the library itself, i.e., you want to tag or place notes on that library. Again this is easily feasible. One need just update the model of the library:
While I don't have an objection to the implementation suggested here, we should definitely take into consideration the W3C annotation model even for these kinds of annotations (called "whole-record", meaning we are annotating the entire document with a tag or some free-text content). See https://www.w3.org/annotation/ and links therein.
I suspect we'll need to collect a lot more information about the annotation itself to be compliant with the model. Given the potential expansion to full-text annotation in ADS and other services, we need to be sure that we can provide an OA-compliant API some day.
We have had several debates, and discussions about tags, and annotations (any many other forms of the same thing). For the foreseeable year, I believe this argument holds true:
So, I hesitantly write the following, and encourage that this feature only be considered if the developers and other team members think it is important.
Tags or Annotations: Backend
Given the way the system is currently architected, it is completely possible to include tags or annotations.
Bibcode level
Bibcode level refers to, annotating a single bibcode
Bib1
with a tag, or an annotation, within a given libraryLib1
.Each user's library contains the following:
The idea would be simple. You can add an annotation to the dictionary of the bibcode, so you'll essentially have a
JSON
blob as:Keep in mind, this would always have a Library scope. If you wanted to use the same tags and annotations for a user through all of our interface, I do not believe the library service is the way to do it, given the way the models are -- but worth thinking about.
Library level
Library level refers to the scope of the library itself, i.e., you want to tag or place notes on that library. Again this is easily feasible. One need just update the model of the library:
Or whatever type of notes you wish to be added to the library at this level. You'd then simply update the REST end points, and migrate the database.
The text was updated successfully, but these errors were encountered: